# Hypernerf Practice
[Huggingface Link](https://huggingface.co/xieyizheng/hypernerf_practice)
This README contains overview on applyiing several custom scenes to reproduce the hypernerf project.
- hand
- dvd
- tomato-mark
## hand
most successful one, trained extensively on V100
interpolation:
![Image 1](gif/hand-interpolate.gif)
novelview:
![Image 2](gif/hand-novelview.gif)
PSNR:
coarse metrics: metric/psnr≈31.7468
fine metrics: metric/psnr≈32.5462
## dvd
not so great
interpolation & novelview:
![Image 3](gif/dvd-interpolate-novelview.gif)
PSNR:
coarse metrics: metric/psnr≈
fine metrics: metric/psnr≈
## tomato-mark
more artifacts
interpolation:
![Image 4](gif/tomato-mark-interpolate.gif)
PSNR:
coarse metrics: metric/psnr≈
fine metrics: metric/psnr≈
# reproduction steps
Due to deprecation, the hypernerf project doesn't work out of the box.
We need code modifications to the environments, dependencies set up and rendering part.
For this reproduction, I only used the notebooks.
The provided instructions below are guaranteed to work.
[Click to reproduction steps](reproduction-instruction.md)
# Dataset:
---
language:
- en
thumbnail:
tags:
license:
datasets:
metrics:
---